Based on a data set collected by GCD from 55 member banks, covering an historical period of 20 years, this paper assesses the empirical level of Credit Conversion Factors (CCF2) for Performance Guarantees or Technical Guarantees as referred in CRR3. The study confirms the average credit conversion factor (CCF) of 10% +/- moc for defaulted customers which validates the case for applying a 20 % CCF in determining exposure at default (EAD), for performance guarantees when calculating Risk Weighted Assets (RWA) for capital purposes.
As part of this updated report, the methodology and the data collection have been enhanced to align with risk modelling practices banks deploy for regulatory capital calculations. This shifts the focus from the total portfolio of all customers to only defaulted customers (in line with regulatory requirements for LGD modelling), and therefore uses a portfolio of defaulted customers and associated payments made under guarantees issued by these defaulted customers to estimate empirical CCFs.
Data and methodology
GCD started collecting historical loss data in 2004, to which member banks have exclusive access. GCD data only covers cases where the borrower has defaulted (using the Basel definition). This database now totals over 302,000 non-retail defaulted loan facilities from around the world. The total GCD defaulted data set is composed of data from the banks who have chosen to be GCD members. These banks’ geographical lending footprint, facilities, and borrower types as well as collateral practices are merged in the database.
In this report GCD bases the analytics on a filtered data set: using specific products, (performance guarantees and financial guarantees) and combining elements of representativeness and data quality. The three facility types that GCD classes as performance guarantees are trade related payment guarantee, other trade related bonds, and trade finance bid or perf bond. The structure of the GCD database reflects the full complexity of the legal relationship between a bank lender and a borrower. Usually, a single company borrower might have multiple types of facilities (revolving loans, term loans, performance guarantee facility etc.). The database is designed to deal with the simplest through to the most complex deals and GCD member banks can access the whole deals structure on facility and obligor levels. For this report, figures are aggregated at facility level.
The methodology uses a portfolio of defaulted customers as the starting point for collecting paid amounts on performance and financial guarantees issued by these defaulted customers. The reference data set is then used to estimate empirical CCF, where the CCF is defined by the following ratio.
The methodology has been applied to the GCD data pool of defaulted customers with performance guarantees facilities (time span 2000-2018). From a business practice perspective, it is important to understand that not all guarantees are claimed for defaulted customers. Further issuing banks often extend claims subject to a mutual agreement between applicant and beneficiary, or do not pay claims if they are discrepant or subject to a legal stay order obtained by the applicant. This often results in a bank not needing to pay out against these guarantees.
The results from this globally representative data pool confirm the low-risk profile of performance guarantees from a credit perspective – reinforcing the rationale for applying a 20 % CCF in determining exposure at default for modelling under the Basel III framework.